This article aims to examine whether Italian has theticity and sentence-focus among its grammatically encoded categories. After the categories of theticity and sentence-focus are introduced, the concepts of grammatically encoded category and category of language use are operationalized along the lines of Integral Linguistics. For a functional category to qualify as a grammatically encoded category of a language, the language should have at least one construction that structurally encodes the category as its indefeasible semantics. The article provides a qualitative investigation of both corpus-based and constructed examples of the eight Italian sentential structures that have hitherto been recognized in the literature as thetic or sentence-focus constructions. It is shown that each of the constructions has information structural uses that are incompatible with a structurally encoded theticity or sentence-focus semantics, i.e., topic-comment structure and narrow focus construal. It is argued that theticity and sentence-focus are therefore categories of language use rather than grammatically encoded categories of Italian.
The categories of theticity and sentence-focus have frequently been used to characterize various linguistic structures in a wide variety of typologically diverse languages (Abraham et al. 2020; Lambrecht 1987, 1994, 2000; Matić 2003; Sasse 1987, 1995, 2006; Ulrich 1985). Also in Italian various sentential structures have been analyzed as thetic and sentence-focus constructions, including the Syntactic Inversion Construction (1), the Existential Construction (2), the Presentational Cleft (3), the avere Cleft (4), the ecco Cleft (5), the Perception Verb Construction (6), the Bare NP Cleft (7), and the Prototypical Cleft (8). These sentential structures can be used to convey utterances that are characterized by a logical thetic judgment, an all-comment structure, and a subject and a predicate that both fall under the scope of the focus operator.
Although the existing literature has clearly shown that Italian has several structures that can be used to express theticity and sentence-focus, it has hitherto remained unexamined whether any of these constructions grammatically encode the categories of theticity and sentence-focus. This article adopts a constructional approach (Fillmore and Kay 1993; Fillmore et al. 1988; Goldberg 1995, 2006; Hoffmann and Trousdale 2013) to study these structures and their functions as form-meaning pairings and draws on Integral Linguistics (Coseriu 1975 , 1987, 1992 , 2007 ; Willems 1997, 2011, 2016; Willems and Munteanu 2021) to establish a distinction between ‘grammatically encoded categories’ and ‘categories of language use’. To qualify as a grammatically encoded category of a language, a functional category needs to be the indefeasible semantics of at least one construction at the lexical, phrasal, clausal, or sentential level of that language. By virtue of being indefeasible, the functional category cannot be absent while the accompanying linguistic form is present, and thus the functional category has a one-to-one corresponding formal counterpart. Categories of language use, on the other hand, capture the various uses that linguistic constructions have, depending on the context, without their presence or absence having a necessary impact on linguistic form.
The present article aims to study whether theticity and sentence-focus are grammatically encoded categories of Italian and therefore investigates whether any of the eight aforementioned Italian constructions encode theticity or sentence-focus as their indefeasible semantics. This research fits within a strand of studies that over the past 25 years have critically examined the functions of several thetic and sentence-focus constructions in various languages. Most studies have focused on specific thetic and sentence-focus constructions, showing that the thetic or sentence-focus reading is often only one possible use among many others, which entails that these functions can be canceled depending on the context (Belligh 2020a; Matić 2003; Karssenberg 2016; Karssenberg and Lahousse 2018; Sasse 1995, 2006). A comprehensive study that aims to deal with all relevant structures in one language to determine whether that language has theticity and sentence-focus among its grammatically encoded categories has hitherto, to our knowledge, been undertaken for Dutch only (Belligh 2020b). For Italian, there have been several studies demonstrating the information structural multifunctionality of some of the Italian thetic and sentence-focus constructions (e.g., Belletti 2018; Bernini 1995; Cardinaletti 2018; Cruschina 2012, 2015; Karssenberg et al. 2017). However, there have hitherto been no comprehensive studies of all previously identified Italian structures that can be used to convey theticity and sentence-focus.
This study aims to fill this gap by providing a qualitative study of the previously identified Italian thetic and sentence-focus constructions. The investigation draws on a number of sources, including reported data in previous studies, corpus-based data and examples constructed on the basis of native-speaker intuition. On the basis of the results, this study aims to gain a better understanding of the eight constructions involved and of the overall status of theticity and sentence-focus in Italian. Additionally, the study aims to contribute to our understanding of the general status of theticity and sentence-focus crosslinguistically, and to the ongoing theoretical debate regarding the relation between grammar and information structure phenomena (cf. Féry 2008; Matić and Wedgwood 2013; Selkirk 2008).
The article is structured as follows. After defining the categories of theticity and sentence-focus (Section 2.1), the article discusses the Italian structures that have hitherto been identified as thetic and sentence-focus constructions (Section 2.2). The article then turns to elucidating the difference between grammatically encoded categories and categories of language use (Section 2.3), elaborates on how information structure phenomena can be studied with regard to questions of grammatical and semantic encoding (Section 2.4) and presents an overview of the relevant research on theticity and sentence-focus with regard to this issue (Section 2.5). In Section 3, the specific methodology applied in this article is presented both with regard to the general strategy used for identifying grammatically encoded categories in languages (Section 3.1) and with regard to the sources of data and the operationalization of the categories of theticity and sentence-focus (Section 3.2). In Section 4, the results of the investigation are first presented for each construction in its own right, after which a general conclusion about Italian is put forward and a number of possible objections to the proposed analysis are addressed. The article ends with a number of conclusions about theticity and sentence-focus in Italian and about the general status of these two functional categories (Section 5).
2 Theoretical background and state of the art
2.1 Theticity and sentence-focus
The functional category of theticity has been defined in two major ways. On the one hand, there is the classical, logical conception that defines theticity as a simple judgment that does not predicate something about some entity or event but merely affirms the existence of that entity or event (Kuroda 1972; Marty 1918; Venier 2002). Thetic judgments therefore attribute something to a situation as a whole and can be formalized as ‘A is’ or ‘A is not’, as in, e.g., (9), (10) and (11). The thetic judgment is defined in contrast with the more predominant categorical judgment, which requires both the recognition of some entity or event and the act of affirming or denying a predication about that entity or event. In this case, there are two subsequent judgments involved, i.e., a judgment of recognition and a judgment of predication, and this can be formalized as ‘A is B’ or ‘A is not B’, as in, e.g., (12).
On the other hand, there is the discourse-based definition of theticity introduced by Sasse (1987). Under this view, theticity is defined as a specific perspective that presents some state of affairs as a nonpredicative unanalyzed whole, as in, e.g., (13). Categorical sentences are reinterpreted in this approach as sentences that express a point of view consisting of a separate “predication base” and a “predicate” (Sasse 1987), as in, e.g., (14). In the terminology of modern information structure studies, Sasse’s distinction between “non-predicative” sentences and “predication base and predicate” sentences can be equated with the difference between “all-comment” sentences and “overt topic constituent-comment” sentences (Gundel 1988 ; Gundel and Fretheim 2004; Lambrecht 1994). One can argue that even in all-comment sentences, a topic-comment relationship is present, with the topic being a nonlinguistic entity located outside of the sentence or, alternatively, a covert linguistic constituent, typically interpreted as pertaining to the spatiotemporal circumstances of the utterance (Dufter and Gabriel 2016; Gundel 1988 ). In any case, even if an external nonlinguistic or covert linguistic topic is considered to be present, no overt topic constituent, i.e., an overt linguistic element functioning as the topic, is present in “all-comment” sentences (Dufter and Gabriel 2016; Krifka 2008; Lambrecht 1994).
The sentence-focus category was proffered by Lambrecht (1987, 1994, 2000, 2001) as one of the three universal types of focus extension distinguished in his theory, in addition to ‘predicate-focus’ and ‘argument-focus’. In the case of predicate-focus construal, the scope of the focus is limited to the predicate, with the subject falling within the scope of the presupposition, as in (15). Alternatively, the scope of the focus can be limited to an argument constituent only, with the predicate constituent falling within the scope of the presupposition, which is labeled argument-focus construal, (see (16)). Third, it is also possible that both the subject and the predicate fall under the scope of the focus operator, which is then said to be an instance of sentence-focus construal, as in, e.g., (17).
The categories of sentence-focus and theticity have frequently been used to characterize many structures in a highly diverse group of languages, including Spanish, French, Romanian, English, German, Japanese, Egyptian Arabic, Fulfulde, Wolof, Buli, Tagalog, Trio, Sikuani, and Jaminjung. Crosslinguistically thetic and sentence-focus constructions can take very diverse forms, ranging from prosodic accentuation of the subject constituent (e.g., in English, German and Turkish), inversion of the morphosyntactic subject and predicate (e.g., in Spanish, Romanian, Albanian, Serbo-Croatian, Russian and Chinese), subject incorporation into the verb (e.g., in Boni) and the use of clefted syntactic structures (e.g., in French and Egyptian Arabic) to the insertion of specific discourse particles (e.g., the joshi –ga in Japanese) (Abraham et al. 2020; Kuroda 1972; Lambrecht 1987, 1994, 2000; Matić 2003; Sasse 1987, 1995, 2006; Ulrich 1985). Furthermore, in many languages, sentence-focus and thetic constructions occur as a group or set of constructions. For example, Dutch has several thetic and sentence-focus constructions, including a Prosodic Inversion Construction, a Syntactic Inversion with Filler Insertion Construction, an Existential Construction, a Non-Prototypical Cleft, and a Perception Verb Construction (Belligh 2018, 2020a, 2020b).
It is important to note that the qualifications for sentence-focus constructionhood are stricter than those for thetic constructionhood. While any linguistic structure that can be used to express theticity tends to be considered a thetic construction in the literature, a linguistic structure is only considered a sentence-focus construction if it is used to convey sentence-focus construal and if the construction also structurally indicates this construal (Lambrecht 1987, 1994, 2000, 2001). For Lambrecht, this structural marking needs to be done in terms of marking the paradigmatic contrast with the typical declarative sentence structure with predicate-focus construal, as in, e.g., (15), which, according to Lambrecht, is the most unmarked type of construction for every language. Displaying sentence-focus construal is thus a necessary but not a sufficient condition for the status of sentence-focus construction. Lambrecht (1987, 1994, 2000 excludes thetic sentences such as (18), which, functionally speaking, clearly have a sentence-focus construal, from the status of sentence-focus construction, because they are structurally indistinguishable from predicate-focus constructions with pronominal subjects that carry both referential and relational presuppositions, as in (19).
Furthermore, sentence-focus constructions are defined exclusively with regard to the focal or nonfocal status of the subject constituent and the predicate constituent. If there are constituents embedded within the predicate constituent that are not in focus, this does not bear on the classification of the sentence as having sentence-focus construal. Consider, for example, sentence (20) from Lambrecht, which serves as an answer to a question about the latest available news on a certain woman.
The clitic direct object at the beginning of the sentence functions as an overt topic constituent and is presupposed and coreferential with the woman who was enquired about in the question. The sentence is therefore clearly categorical rather than thetic. Nonetheless, this sentence is classified as a sentence-focus construction by Lambrecht since both the subject (il marito) and the predicate constituent as a whole are in focus. The nonfocal direct object (l’), which is only part of the broader predicate constituent (l’ha lasciata), has no bearing on the classification of the sentence as a sentence-focus construction because it is not the subject nor does it constitute the entire predicate constituent.
2.2 Italian thetic and sentence-focus constructions
2.2.1 Preliminary remarks on constructionhood and the constructions involved
In Italian, various linguistic structures at the sentential level have been analyzed as “thetic and sentence-focus constructions”. Given that the category of theticity has been conceptualized in two ways and that the related category of sentence-focus is not identical to theticity, questions might arise as to the utility of considering Italian thetic and sentence-focus constructions as a single group of constructions. However, it has been demonstrated that logical theticity and discourse-based theticity can be used to identify the same set of constructions and that discourse-based and logical theticity should therefore be seen as two ways of approaching one and the same category rather than two different but related categories (Belligh 2020a). Although theticity and sentence-focus are not identical categories and therefore do not necessarily relate to the same set of constructions (cf. Sections 2.1 and 3.2), the same set of Italian structures has been identified in the existing literature using theticity or sentence-focus as the onomasiological point of departure.
In this article, we approach the Italian sentential structures that can be used to express theticity and sentence-focus from a constructional point of view. This entails that, along the lines of Construction Grammar (Fillmore and Kay 1993; Fillmore et al. 1988; Goldberg 1995, 2006; Hoffmann and Trousdale 2013), we study these structures and their information structural functions as form-meaning pairings at the sentential level. From a constructional point of view, information structure constructions can be seen as pairings of a procedural, nonconceptual meaning and some kind of form, i.e., a sentential pattern defined by its morphosyntactic and prosodic characteristics (Goldberg 2006; Lambrecht 1994, 2000, 2001; Leino 2013). The original criterion to define a construction adopted in Construction Grammar stipulates that some aspects of the meaning or the form of the construction should not be compositionally derivable from the meaning or form of the component parts (Fillmore and Kay 1993; Goldberg 1995). Under this view, information structure constructions definitely qualify as constructions, given that information structural meaning—insofar as relational governess, i.e., topic-comment structure and focus-background articulation, is concerned—occurs only at the clausal level and cannot be reduced to the referential givenness or newness of the constituents involved (Gundel 1988 ; Gundel and Fretheim 2004; Lambrecht 1994).
Terminologically we refer to all Italian sentential structures that have been found in previous research to convey theticity and sentence-focus as “thetic and sentence-focus constructions”. However, one of the main goals of our study is to examine whether any of these structures are actually structurally dedicated to the expression of theticity, sentence-focus, or both and thus constitute a form-meaning pairing in the specific sense of Integral Linguistics (cf. Section 2.4). By enriching Construction Grammar with the conceptual distinction between grammatically encoded meanings and meanings that pertain to (normal) language use, this study can be situated in a recent strand of studies that aim to integrate several insights from Integral Linguistics in the Construction Grammar framework (cf. Belligh 2020a; Coene 2006; Coene and Willems 2006; De Vaere et al. 2020). While we do not subscribe to all tenets of any of the specific schools within Construction Grammar, we adopt a moderate approach to Construction Grammar. In particular, we assume that in addition to constructions, grammatical rules proper are needed to understand language, along the lines of Berkeley Construction Grammar (Fillmore and Kay 1993; Fillmore et al. 1988) and the Parallel Framework (Jackendoff 2010), rather than subscribing to the view that constructions are the only category needed to understand language, as maintained, for example, in Cognitive Construction Grammar (Goldberg 2006).
Regarding the form of the sentential constructions under investigation, several remarks are in place. First, there is considerable formal heterogeneity among the Italian structures involved. In particular, some of the sentential structures under investigation are monoclausal in nature, whereas others are biclausal. Although from a purely syntactic point of view, this is a substantial difference, this is not necessarily problematic from a constructional point of view. In fact, constructional analyses of sentential structures and their information structural functions have been applied to monoclausal and biclausal structures alike (e.g., Lambrecht 1994, 2000, 2001). Second, in line with the main tenets of Construction Grammar, we maintain that the form of constructions can be determined by various possible means and at various possible levels, including phonology, morphology, syntax and prosody (Fillmore and Kay 1993; Goldberg 2006; Hoffman and Trousdale 2013). Given that we are dealing with constructions at the sentential level in this study, the focus of the present article is at the level of syntax. Various syntactic characteristics, pertaining both to linear syntax and to structural syntactic relations, are taken to be able to distinguish one construction from other constructions. At the same time, it must be noted that in constructional, nonderivational approaches to grammar (Fillmore and Kay 1993; Goldberg 1995, 2006; Hoffmann and Trousdale 2013), syntactic characteristics that merely pertain to the theoretical syntactic movements typical of Generative Grammar are usually not taken into consideration to determine constructionhood. Third, within the limits of this article, we do not delve deeply into the prosodic characteristics of the constructions involved. While in empirical practice the dominant focus in constructional approaches is on morphosyntax rather than prosody, letting prosody differentiate between several constructions is in principle entirely consistent with the main tenets of constructional thought and has also been applied in the study of several information structure constructions (cf. Lambrecht 1994, 2000). Not dealing in great detail with the prosodic characteristics of the constructions involved might therefore constitute a major limitation of our approach, which we discuss further in the article (Sections 4.9 and 5).
A final terminological complication that needs to be taken into account is that in the Italian literature thetic and sentence-focus constructions have often been analyzed as strutture/frasi/costruzioni presentative ‘presentational constructions’, rather than as thetic or sentence-focus constructions. This is often the case in studies that are interested in the function of these constructions to introduce referentially new entities and events into the discourse world rather than in their topic-comment structure or focus-background construal. However, in the research tradition that describes these constructions as “presentational”, it is also frequently acknowledged that their logical structure, topic-comment articulation and focus-background construal actually correspond to those captured by the categories of theticity and sentence-focus (Berretta 1995; Cruschina 2012, 2015, 2018; De Cesare 2007; Marzo and Crocco 2015; Meulleman 2012; Venier 2002).
In what follows, an overview is provided of the eight Italian structures that have hitherto been identified as thetic and sentence-focus constructions. Examples that illustrate the thetic and sentence-focus uses of these constructions are taken from the existing literature (which is indicated by reference to a specific article), constructed on the basis of native-speaker intuition (without specific indication) or retrieved from the LIP corpus (which is indicated with their specific LIP code). The LIP corpus is a collection of spoken Italian collected in naturalistic settings in four different Italian cities, namely, Milan, Florence, Rome and Naples. Compared to many present-day web-based corpora of written language, the LIP is a small corpus, as it consists of 469 texts, corresponding to approximately 490,000 words. However, it is still considered one of the most valuable language resources available for Italian, as it both represents the most relevant Italian regional varieties and provides spontaneous spoken data of different sorts. Construction tokens were retrieved from the LIP through the search engine made available by the University of Graz (Bellini and Schneider 2003–2019).
2.2.2 Syntactic Inversion Construction
The Italian Syntactic Inversion Construction (SIC) is a broad clausal template formally characterized by its specific linear syntax, whereby the construction merely stipulates that the morphosyntactic subject should follow the inflected verb within the same nuclear clause, e.g., (21), rather than preceding it as in subject-verb constructions like (22).
It has been argued that the SIC occurs mostly with verbs that indicate presence, absence, appearing, disappearing, continuation, and change and verbs that show “semantic solidarity” with the postverbal subject (Bernini 1995; Wandruszka 1982; Wehr 2012). Furthermore, a strong correlation—albeit a far from absolute one—has been found between the occurrence of a postverbal subject and the class of unaccusative verbs (Bernini 1995; Sornicola 1995). In fact, many unergative verbs, as in, e.g., (23), and, in specific contexts, transitive verbs, as in, e.g., (24), can also occur with the SIC (Belletti 2004, 2005, 2018; Berretta 1995; Fiorentino 2005; Meulleman 2012; Venier 2002). With transitive verbs, the inverted word order is more easily found if it is preceded by a left dislocated direct object or an adverbial phrase of place or time (Belletti 2005). Furthermore, there is some correlation between the occurrence of postverbal subjects and syntactically heavy subjects (Cennamo 1995), e.g., (25).
The SIC has been widely recognized as one of the most predominant Italian thetic and sentence-focus constructions (Belletti 2004, 2018; Bernini 1995; Berretta 1995; Bonvino 2005; Cardinaletti 2018; Cennamo 1995; De Cesare 2007; Fiorentino 2005; Meulleman 2012; Sornicola 1995; Venier 2002; Wandruszka 1982; Wehr 2012) and is often found with these uses, as in, e.g., Examples (26), (27) and (28).
2.2.3 Existential Construction
The Italian Existential Construction (EC) is formally characterized by the presence of the expletive pronominal ci, the copula essere and a postverbal pivot NP, as in (29), and can optionally be followed by a locative coda, as in (30). There is extensive literature distinguishing the Existential Construction from other related sentential constructions introduced by ci and the copula essere, in particular from locative constructions, which can be further divided into deictic locatives and inverse locatives (Berruto 1986a; Cruschina 2012, 2015, 2018; De Cesare 2007).
It has been argued that the difference between locatives and existentials proper resides in the following three characteristics. First, the presence of a definite NP in the pivot position is connected to the occurrence of a locative construction. Second, the expletive pronominal ci functions as a pro-argument in the EC, whereas it functions as a pro-predicate in locatives. Third, the status of the postverbal noun phrase is a predicate in existentials, whereas it is an argument in locatives (Cruschina 2012, 2015, 2018). The EC has been widely recognized as a thetic and sentence-focus construction (Berruto 1986a; Cruschina 2012, 2015, 2018; De Cesare 2007; Meulleman 2012) and frequently occurs with these kinds of information structural properties, as in, e.g., Examples (31) and (32).
2.2.4 Presentational Cleft
The Italian Presentational Cleft (PC), in the Italian literature referred to as c’è presentativo, or occasionally as c’è focalizzante (De Cesare 2007), illustrated in (33) and (34), is another type within the group of Italian sentential constructions introduced by the expletive pronominal ci and the copula essere. Although the PC shares many traits with the EC, notably its introduction by ci and the copula essere ‘to be’, it has been commonplace to distinguish the two constructions from each other and from other constructions within the group, such as locative constructions (Berretta 1995; Berruto 1986a; Cruschina 2012, 2015, 2018; De Cesare 2007; Marzo and Crocco 2015).
Two traits have been proposed as defining criteria for the PC. First, at the semantic and/or pragmatic level, it has been argued that the ci of the PC would be semantically empty, while this would not be the case for the ci in the Italian existential and locative constructions (Cruschina 2012, 2015, 2018; De Cesare 2007). It often cannot be attributed a role in the predication; it does not stand for the argument, as in the EC, or for the predicate, as in locative constructions. Second, the obligatory presence of an additional constituent after the postverbal NP has been proposed as the defining trait par excellence for this construction (Berretta 1995; Cruschina 2012, 2015, 2018; De Cesare 2007). This obligatory constituent can be an adjectival or prepositional phrase or, more commonly, a subordinate clause (Cruschina 2012, 2015, 2018; De Cesare 2007). In the latter case, the presence of a subordinate clause, and hence the biclausal nature of the construction, is a necessary component of instances of this construction, whereas it is optional for other types of Italian ci clauses (Cruschina 2012, 2015, 2018; De Cesare 2007). Furthermore, it has been noted that definite NPs can easily appear in the PC but not in the EC, which further distinguishes the PC from the EC proper (Cruschina 2018).
Additionally, it has been argued that the definition of the PC should be narrowed down further so that it can contain only a specific subtype of subordinate clause, i.e., the pseudorelative clause. Pseudorelative clauses are distinguished from two kinds of proper relative clauses, viz., restrictive relative clauses (relativa restrittiva) and appositive relative clauses (relativa appositiva), which are said to occur optionally with other ci sentences, such as with locatives and with the EC (Cruschina 2012, 2015, 2018). Pseudorelative clauses differ from proper relatives with regard to a number of specific syntactic properties (Radford 1975). In the case of Italian (Casalicchio 2013; Cinque 1988), pseudorelative clauses are taken to be identified based on the following set of features: the antecedent of the subordinate clause can be a proper noun; the complementizer che cannot be replaced by il quale; the complementizer can only be the syntactic subject of the subordinate clause; the events described in the main clause and in the pseudorelative clause take place simultaneously; and the pseudorelative clause usually has a stage-level predicate.
The PC has been widely recognized as a predominant Italian thetic and sentence-focus construction (Berretta 1995; Berruto 1986a; Cruschina 2012, 2015, 2018; De Cesare 2007; Fiorentino 2005; Marzo and Crocco 2015; Meulleman 2012; Venier 2002) and is often found with these uses, as in, e.g., Examples (33)–(36).
2.2.5 Avere Cleft
The avere Cleft (AC) is a biclausal sentential construction formally defined by its introduction by an inflected form of avere ‘to have’ in the first-person singular or plural, followed by a direct object NP, which in turn functions as the antecedent of the syntactic subject of an ensuing pseudorelative clause (Berretta 1995; Fiorentino 2005; Venier 2002), as illustrated in Examples (37) and (38). As a member of the cleft family, the AC shows structural similarities to the PC, the ecco Cleft, the Bare NP Cleft and the Prototypical Cleft (Berretta 1995; Roggia 2009; Venier 2002). The AC has been recognized as a thetic and sentence-focus in the work of Berretta (1995), Venier (2002), and Fiorentino (2005).
It has been noted that the avere Cleft is more often used than the Presentational Cleft when there is some (weak) link of possession between the subject of the main clause and the direct object NP, whereas the PC tends to be used when this is not the case (Berretta 1995). This can be interpreted as an indication that the inflected form of avere at the beginning of the AC is less grammaticalized than the highly grammaticalized ci essere that appears at the beginning of the PC.
2.2.6 Ecco Cleft
The Ecco Cleft (EccoC) is formally characterized by its clefted syntactic structure, which is introduced by the Italian adverb ecco followed by a noun phrase, which in turn functions as the antecedent of the syntactic subject of an ensuing pseudorelative clause (Berretta 1995), as in Examples (39) and (40). As a member of the cleft family, the EccoC shows structural similarities to the PC, AC, bare NP Cleft and the Prototypical Cleft (Berretta 1995; Roggia 2009). The Ecco Cleft has, to the best of our knowledge, been recognized as a thetic and sentence-focus construction only in the work of Berretta (1995).
2.2.7 Perception Verb Construction
The Italian Perception Verb Construction (PCV) is a biclausal sentential construction that consists of a first clause with an inflected transitive perception verb in the first-person singular or plural, followed by a direct object NP that functions as the antecedent of the syntactic subject of an ensuing pseudorelative clause (Cruschina 2012), as in, e.g., Examples (41), (42) and (43). The range of available perception verbs includes verbs such as vedere ‘to see’ and sentire ‘to hear’ and ‘to smell’. Although the PCV has not been widely discussed as an Italian thetic and sentence-focus construction, it has been recognized as such by Berretta (1995).
2.2.8 Bare NP Cleft
The Italian Bare NP Cleft (BNPC) is a construction that consists of a bare NP that is introduced in the discourse world as such, followed by a (pseudo)relative clause, as in Examples (44) and (45). As a member of the cleft family, the BNPC shows structural similarities to the PC, AC, EccoC and Prototypical Cleft (Berretta 1995; Roggia 2009). It is a severely understudied construction that has been recognized as a thetic and sentence-focus only in the work of Berretta (1995).
2.2.9 Prototypical Cleft
The Italian Prototypical Cleft (PrC) is a biclausal sentential construction introduced by an inflected form of the verb essere ‘to be’ in the third-person singular or plural, followed by an NP, in turn followed either by a relative appositive clause or by a pseudorelative clause, as in, e.g., (46) and (47). As a member of the cleft family, the PrC shows structural similarities to the PC, AC, EccoC and BNPC (Berretta 1995; Roggia 2009). Although the Prototypical Cleft is commonly associated with argument-focus construal both in Italian and crosslinguistically (Lambrecht 1994, 2001; Roggia 2009), the PrC has also been recognized as an Italian construction that can be used to convey the information structural categories of theticity and sentence-focus by Berretta (1995) and Roggia (2009).
2.3 Grammatically encoded categories versus categories of language use
In Italian and in several other languages, many linguistic structures can be characterized by a thetic or sentence-focus reading (Sections 2.1 and 2.2). However, it is one thing to show that a certain linguistic structure can be used for a certain function, but it is quite another to show that this specific function is grammatically encoded by the same linguistic structure. The fact that a linguistic construction can be used by language users to convey a certain functional category does not entail that that particular functional category is grammatically encoded by that construction. Language users can convey all kinds of functions by means of various linguistic structures, but only some functions are unambiguously grammatically encoded by a dedicated structure in specific natural languages.
To clarify the difference between grammatically encoded categories and categories of language use, we rely on the account of this distinction developed within the framework of Integral Linguistics (IL) (Coseriu 1975 , 1987, 1992 , 2007 ; Willems 1997, 2011, 2016; Willems and Munteanu 2021). Although we approach the distinction from the specific outlook of Integral Linguistics, partly similar views can also be found in other theoretical frameworks, most notably in Neo-Gricean Pragmatics (cf. Belligh and Willems 2021 for a discussion). In fact, many frameworks have developed accounts of the grammar-semantics-pragmatics distinction and put forward proposals on how to distinguish between the specific meanings and functions that are structurally encoded in the lexicon and grammar of a specific natural language and other facets of meaning that accrue to expressions when people put language to use. In comparison with those of other frameworks, the specific characteristics of the model of the grammar-semantics-pragmatics interface put forward by IL lie in its focus on the paradigmatically delimited, language-specific nature of grammatically encoded semantics, its strict adherence to the principle of defeasibility to distinguish grammatically encoded meaning from pragmatically inferred meaning in a principled way, and its application of the distinction between grammatically encoded meaning and other aspects of meaning to both lexical semantics and the functions of grammatical structures (Belligh and Willems 2021). IL has furthermore developed a fully worked-out account regarding the relation between the study of the semantics and pragmatics of specific constructions and questions pertaining to the overall status of specific functional categories in linguistic systems taken as a whole.
From an IL perspective, both grammatically encoded categories and categories of language use capture important aspects of the meaning of constructions. Categories of language use describe the various pragmatic uses that linguistic constructions can have, depending on the context. Grammatically encoded categories, on the other hand, capture the grammatically encoded semantics of constructions that underlie all pragmatic uses in context. The crucial difference for IL is that only grammatically encoded categories correspond strictly in a one-to-one fashion to specific linguistic forms. Only these categories and their coextensive forms constitute form-meaning pairings that can be considered Saussurean linguistic signs proper. In fact, a linguistic form almost always has different pragmatic uses, depending on the context. Thus, it is possible to cancel every single one of these pragmatic uses by changing the context without this having any impact on the occurrence or nonoccurrence of the linguistic form involved. A grammatically encoded category, on the other hand, captures the structurally encoded semantics of a linguistic construction and is not cancelable without an accompanying change in linguistic form. A crucial trait of grammatically encoded categories is therefore that they are indefeasible, viz., that there are no contexts in which the linguistic form occurs in which the encoded meaning or function would be canceled. Due to their abstract nature, these categories allow us to understand language by offering a firm bedrock for in-depth interpretative explanations. The specific pragmatic senses attested in language usage can be conveyed by language users on the basis of the combination of the knowledge of the structurally encoded semantics, world knowledge, knowledge of discourse traditions, and general inferential capacities. Consider, for example, the English preposition over. It has various senses in usage, as in Examples (48), (49), (50) and (51) (cf. Van der Gucht et al. 2007).
While each of these senses aptly characterizes one of the pragmatic uses of the preposition, they all remain cancelable, depending on the changing context. On the other hand, the following, indefeasible, structurally encoded meaning has been proposed for over: ‘positioning of X vis-à-vis a reference point Y which is inferior to X’ (Van der Gucht et al. 2007). This structurally encoded function underlies the various pragmatic uses of the preposition and cannot be canceled by changing the context. If one changes aspects of the structurally encoded meaning, e.g., from ‘inferior’ to ‘superior’, there is an accompanying change in linguistic form, resulting in the use of the preposition under rather than over. In a similar vein, the English preposition to has various uses, as illustrated in sentences (52), (53), (54), and (55) (De Cuypere 2013). Each of these functions is, however, cancelable depending on changing contexts. The grammatically encoded, indefeasible meaning of the preposition, on the other hand, has been described as ‘establisher of relationship between X and reference point Y’ (De Cuypere 2013).
It is worthwhile to emphasize that, from an IL perspective, the functional categories that are not grammatically encoded in specific languages but bear on language use are more than mere “comparative concepts” (cf. Haspelmath 2010) that are only useful for comparing languages in the meta-language of linguistics. If the functional categories that describe language use are aptly chosen and well developed, they characterize important aspects of real language use that go beyond the limited set of functions and meanings that are grammatically encoded in particular language systems. Additionally, due to the constant effects of language change, categories of language use can become grammatically encoded categories over time. However, from a synchronic point of view, which by definition entails the study of a language at only one point in time, there is a qualitative difference between the two kinds of functional categories.
2.4 Grammatically encoded categories, categories of language use and information structure
The previous two examples of IL analyses relate to semantics and pragmatics at the lexical level. However, as stressed earlier, within IL, the distinction between structurally encoded meaning and inferred meaning is taken to apply more broadly to the study of grammatical functions as well. Within IL, it is furthermore maintained that the meanings and functions that are found at the level of clausal and sentential constructions, which are one of the central objects of study in Construction Grammar, should also be subjected to the question of what is grammatically encoded by the constructions involved and what, on the other hand, belongs to the level of pragmatics (Coene 2006; Coene and Willems 2006). It has been argued that several meanings that have been identified in constructional analyses as belonging to the semantics of clausal and sentential constructions are in fact instances of meaning at the level of pragmatics rather than grammatically encoded semantics (Belligh 2020a; Coene and Willems 2006).
Some of the most important meanings or functions that are found at the level of clausal and sentential constructions relate to the domain of information structure. In fact, from a constructional point of view, linguistic analyses in the field of information structure link some kind of meaning, mostly meta-linguistically described in procedural and non-truth-conditional terms, to some kind of form, i.e., a prosodic or syntactic sentential pattern (Goldberg 2006; Lambrecht 1994, 2000, 2001; Leino 2013). However, in many linguistic frameworks, information structure is allocated to a separate tier in the model, distinct from “semantics” (e.g., Halliday and Mathiessen 2004; Jackendoff 2010; Lambrecht 1994). While carving up the broad spectrum of meaning is of course methodologically sound, since one can surely make meaningful distinctions based on what kind of content is communicated, these divisions should not obscure the fact that information structure is part and parcel of linguistic meaning. Basic notions of information structure, such as topic, comment and focus, are by nature content-based; viz., they belong to the meaning side of constructions, not to the formal side. Given that information structure is ultimately a kind of meaning, it should in principle be targeted with the same questions as other aspects of linguistic meaning.
Although the distinction between structurally encoded semantics and inferred pragmatics has hitherto been mostly applied to constructions conveying truth-conditional kinds of meaning, it is entirely valid to pursue the question of how much information structural meaning is structurally encoded and how much is pragmatically inferred (Ariel 2010; Belligh 2020a; Belligh and Willems 2021; De Vaere et al. 2020). For information structural functions to count as structurally encoded, linguistic structures need to be linked one-to-one to specific information structural functions and settings, e.g., constructions specifically dedicated to the expression of argument-focus, while not occurring with other information structural functions or in other settings. Finding what kind of information structural configurations go hand in hand with specific constructions is considered by Lambrecht (1987) and many other scholars in the field as the goal of information structure research proper:
Pragmatic structure without corresponding grammatical structure cannot be captured with rules of grammar and lies therefore outside the domain of linguistics proper. (Lambrecht 1987: 373) [emphasis mine]
A point that might create some terminological confusion is that the semantics-pragmatics distinction has also been defined on the basis of the kind of meaning involved. In particular, it has been defined as the distinction between truth-conditional and non-truth-conditional meaning, often in the hope that this distinction would align with the distinction between structurally encoded and inferred meaning. However, closer scrutiny has revealed that these two distinctions are not coextensive but in fact entirely orthogonal. It has therefore been argued that both truth-conditional and non-truth-conditional meaning, including information structural meaning, need to be studied with regard to the grammatically encoded or inferred status of the meanings involved (cf. Ariel 2010 for a discussion).
One of the foundational notions of classical information structure research is the idea that information structural categories are grammatically encoded in the grammars of specific languages (Krifka 2008) and that, for example, dedicated predicate-focus constructions, argument-focus constructions and sentence-focus constructions can be determined (Lambrecht 1987, 1994, 2000, 2001; Leino 2013). This foundational notion has recently been increasingly challenged. For example, Selkirk (2008) argued that in the context of the Generative Grammar framework, grammatical structure should be seen as encoding ‘contrastive focus’ and ‘discourse-givenness’, but not general ‘informational focus’, which under this view has no place in sentence grammar. Féry (2008) argues that information structure categories are not structurally encoded on a one-to-one basis by specific linguistic structures, which in her view serve as mere cues for the hearers to sort out which information structural roles have to be assigned. Matić and Wedgwood (2013) propose a critical account of the information structural notion of focus, showing on a typological scale that upon closer scrutiny, all kinds of focus structures in a great variety of languages cannot be considered to be dedicated to the expression of a single universal notion of focus. This led Matić and Wedgwood (2013: 127) to conclude that focus is not a proper linguistic category but should be considered “as a heuristic tool” and that it should be merely employed “as a means of identifying structural patterns that languages use to generate a certain number of related pragmatic effects”, along the lines of Haspelmath’s notion of a (2010) “comparative concept”.
2.5 The status of the categories of theticity and sentence-focus
In the existing literature on theticity, various indications can be found that are relevant for assessing the status of this functional category (cf. Belligh 2020b). In the seminal logical work of Marty (1918), thetic and categorical judgments are logical categories of thinking that can be expressed by linguistic means without there necessarily being linguistic structures unambiguously encoding these types of judgment (cf. Sornicola 1995; Ulrich 1985; Venier 2002). Likewise, in the seminal linguistic work of Kuroda (1972) on Japanese, there are already indications that one and the same linguistic structure, in particular the joshi -ga, can be used to express both thetic judgments and judgment material without a specific judgment form, which indicates the joshi’s nondedicated character.
In more modern work on theticity, it is stressed that thetic constructions are often “polysemous”; viz., they exhibit various other information structural uses, such as narrow focus construal on the verb and contrastive argument-focus construal (Matić 2003; Sasse 1995, 2006). Theticity is then sometimes conceptualized as an emerging interpretative consequence of specific linguistic constructions rather than as a grammatically encoded function (Matić 2003; Rosengren 1997; Sasse 2006). A similar position has been defended by Lambrecht (1987, 1994, 2000), who claims that theticity is not always reflected in a corresponding grammatical form. In fact, from a conceptual perspective, (56) is as thetic as (57). By contrast, from a structural perspective, (57) stands in a paradigmatic contrast, here prosody based, with the overt topic constituent-comment sentence My neck HURTS, whereas (56) does not.
Instead, Lambrecht (1987, 1994, 2000) proposes sentence-focus as the real grammatically encoded function of those thetic constructions that are formally distinguishable from predicate-focus sentences. However, in numerous cases, sentence-focus constructions also exhibit other kinds of focus construal, a finding that could undermine the idea of sentence-focus as the grammatically encoded function of sentence-focus constructions. Since the various categories of focus construal are defined paradigmatically in contrast to one another, it is hard to explain how a construction with one type of focus construal as its grammatically encoded function could allow for other kinds of focus construal as derived uses. When confronted with this difficulty, Lambrecht (1987, 1994, 2000 resorts to the idea of homonymy. Lambrecht readily admits that there is often ambiguity between argument-focus and sentence-focus construal, as demonstrated in (58) and (59), but he considers this ambiguity to be due not to an underspecified grammatically encoded meaning or function underlying both sentence-focus and argument-focus construal or to a genuine case of grammatically encoded polysemy but rather to the homonymy of sentence-focus and argument-focus constructions.
Furthermore, predicate-focus constructions are often found to exhibit sentence-focus construal, which Lambrecht tries to explain as being a consequence of the “neutral” or unmarked position of predicate-focus constructions in the language system, rather than by invoking additional homonymy. Lambrecht (1987) furthermore maintains that a predicate-focus construction can be used for sentence-focus construal only if it is without a full NP subject. To account for the structural blurring between sentence-focus and predicate-focus construal in pronominal sentences, as in, e.g., Example (60), Lambrecht adds the proviso that the presence of a full lexical subject NP is needed for a construction to qualify as a sentence-focus construction. If a speaker wants to use a sentence with a full NP subject and sentence-focus construal, he/she is forced to use a full-fledged and formally marked sentence-focus construction. Lambrecht’s theory of focus types is thus characterized by a fair share of homonymy and ambiguity. The only type of focus ambiguity that is explicitly excluded as a possibility in his analyses is a full-fledged sentence-focus construction exhibiting predicate-focus construal (Lambrecht 1987: 375).
Last, some recent studies have challenged the dedicatedness of constructions analyzed as both thetic and sentence-focus constructions. Deguchi (2012) argues that the Japanese joshi –ga and –wa, the constructions also discussed in the seminal work of Kuroda (1972), do not “uniformly represent” the thetic and the categorical judgment, respectively. The French il y a cleft, a prominent thetic and sentence-focus construction, has been shown to have various information structural uses that are radically different from theticity and sentence-focus (Karssenberg 2016; Karssenberg and Lahousse 2018). The Dutch Syntactic Inversion Construction introduced by er was shown to have uses that are incompatible with the hypothesis of a structurally encoded theticity or sentence-focus semantics (Belligh 2020a). Furthermore, it has been argued that this is not only the case for this specific Dutch construction but also for all Dutch constructions that have hitherto been identified as thetic or sentence-focus constructions (Belligh 2020b).
There have hitherto been no comprehensive studies of all previously identified Italian constructions that can convey theticity and sentence-focus with a view to determining whether Italian as a language system has theticity and/or sentence-focus among its grammatically encoded categories. Several studies have demonstrated the information structural multifunctionality of the SIC (Bernini 1995; Berretta 1995; Cardinaletti 2018; Meulleman 2012; Sornicola 1995; Wandruszka 1982), the EC (Cruschina 2012, 2015, 2018), and the PC (Karssenberg et al. 2017), but the other five constructions have, to our knowledge, not yet been addressed by this line of research.
3.1 Identifying theticity and sentence-focus as grammatically encoded categories
Following the theoretical model put forward by IL, we maintain that for a category to qualify as a full-fledged grammatically encoded category of a given language, the language needs to have at least one construction, at the lexical, phrasal, clausal, or sentential level, that unambiguously encodes the functional category as its indefeasible semantics. To examine whether a certain functional category is a grammatically encoded category in a language system, it is therefore necessary to study the semantics and pragmatics of all relevant structures that could potentially grammatically encode the function under scrutiny. To achieve this, a number of subsequent research steps need to be taken.
By taking the functional category under scrutiny as the onomasiological point of departure, we first need to identify the various structures in a language that can be used to convey this particular category. Once all relevant structures have been identified, we need to turn to examining each of them in their own right from a semasiological point of view. For the investigation of the status of theticity and sentence-focus in Italian, the onomasiological phase of the research is considered to have been already sufficiently dealt with in the existing literature. Our research therefore builds on the assumption that the wide array of previous studies on theticity and sentence-focus in Italian have identified the relevant constructions that can be used to convey (one of) the two functional categories. The eight thetic and sentence-focus constructions discussed in Section 2.2 are therefore taken as the constructions of interest that possibly encode theticity or sentence-focus in a strictly grammatical fashion.
In the second step of the research, the identified constructions need to be the object of investigation of a semasiological study. For the semasiological aspect of the study, it is necessary to study the various functions of every construction that are attested in language use. This allows us to determine whether all of the attested uses are compatible with the functional category hypothesized as the grammatically encoded meaning or function of the construction. For the hypothesis to hold, it needs to be plausible that language users can construct all attested uses out of the hypothesized grammatically encoded meaning by relying on general world knowledge, knowledge of discourse traditions, and general inferential capacities. If this is the case for every possible use of the construction, it can be concluded that the functional category under investigation is grammatically encoded by a dedicated linguistic structure and that the category therefore is grammatically encoded in the language system. If, on the other hand, the various attested uses cannot be reconciled with the hypothesis of the functional category as the structurally encoded meaning, it is indicative of the not grammatically encoded status of the category in the language under scrutiny. Furthermore, it must not be possible to change the hypothesized grammatically encoded function or meaning by, for example, changing the context connected to that function without resulting in any change in the corresponding linguistic form.
Rather than charting the complete usage potentials of the eight Italian thetic and sentence-focus constructions for their own sake, our study focuses on specific uses of the constructions that are relevant for falsifying the hypothesis that any of these constructions grammatically encode theticity or sentence-focus. If fact, if a construction can be used to convey meanings or fulfill functions that are incompatible with a hypothetical grammatically encoded theticity or sentence-focus semantics, the construction cannot be said to encode theticity or sentence-focus in a strictly grammatical fashion. If a construction is found with a categorical overt topic constituent-comment structure, rather than a thetic all-comment structure, then theticity cannot be considered the structurally encoded meaning of the construction because the absence of an overt topic constituent is a logical prerequisite for a thetic all-comment structure. If a construction is found to have an argument-focus or predicate-focus construal, then sentence-focus cannot be the structurally encoded meaning of the construction as the three types of focus construal are defined in mutually exclusive terms with regard to one another.
The reason for adopting the aforementioned criteria is twofold. First, given that theticity, categoricality and the different kinds of focus construal are established negatively in relation to one another, it is highly implausible that language users, using world knowledge, knowledge of discourse traditions, and general inferential processes, would get from a grammatically encoded category to a category that is established negatively in relation to the former. Second, if it is possible for one linguistic structure to have two radically opposed functions depending on the context, it follows that each of these two functions can be canceled without having any impact on grammatical form. Given that interdependence of form and function is indispensable for qualification as a grammatically encoded category (Section 2.3), these functions cannot be considered to be grammatically encoded in the specific language under investigation.
3.2 Sources of data and the operationalization of the categories of theticity and sentence-focus
3.2.1 Sources of data
To find a sufficient variety of possible uses of the eight constructions under investigation, the study draws on various sources, including data extracted from the LIP Corpus, data reported in previous studies, and examples constructed on the basis of native-speaker intuition. Construction tokens of the eight construction types involved were extracted by searching through the LIP Corpus (cf. Section 2.2 for a brief discussion of the corpus). The examples were extracted by using the search tool of the LIP corpus (http://badip.uni-graz.at/it/), which allows simple queries for words and parts of speech. Since some of the constructions examined in this work appear only seldom in written language (Marzo and Crocco 2015), we relied on a corpus of spoken Italian to collect the examples. A quantitative analysis of the frequency of the relevant constructions in present-day Italian goes beyond the scope and aims of this paper, so we limit ourselves to a qualitative investigation of the possible uses of the constructions. The retrieved corpus tokens were annotated by means of a qualitative analysis for the relevant information structural factors that pertain to the categories of theticity and sentence-focus. The examples constructed on the basis of native-speaker intuition were used in our analysis to illustrate further possible uses of the constructions not directly attested in the corpus material. One of the authors of the article used her native-speaker intuitive knowledge of Italian to construct a number of examples with some context to demonstrate those possible uses.
3.2.2 Operationalizing the categories of theticity and sentence-focus
Given the functional equivalence between the logical definition of theticity and the discourse-based definition of theticity (Section 2.2.1), we decided to code only for the discourse-based approach and leave out the logical approach for our empirical investigation. Theticity and sentence-focus as functional categories, on the other hand, cannot be collapsed for two reasons. The first reason is that topic-comment structure and focus-background articulation are two levels of information structure that are similar, insofar as they both relate to relational givenness (Gundel and Fretheim 2004), but not identical. In fact, it has been demonstrated that the two dimensions cannot be completely reduced to one another (Cruschina 2012; Dufter and Gabriel 2016; Lambrecht 1994). Second, as discussed in Section 2.1, sentence-focus as a category has an inherent structural component, since it is defined on the basis of the focused or backgrounded status of specific constituents, whereas this kind of criterion is absent from the definition of theticity as a category. This has the consequence that certain sentences can be sentence-focus constructions without being thetic, like (20), while others can be thetic without being sentence-focus constructions, like (18).
From the adoption of the discourse-based approach to theticity, it follows that the topic-comment structure of sentences is crucial to determining their relation to the functional category theticity. The key notion we used to assess topic-comment structure is the interpretative notion of ‘aboutness’ (Gundel 1988 ; Gundel and Fretheim 2004; Lambrecht 1994). For the crucial notions of topic, topic constituent and comment, the following working definitions were adopted for our analysis, with inspiration from Lambrecht (1994) and Gundel and Fretheim (2004).
A referent is interpreted as the topic of a proposition if in a given situation the proposition is construed as being about this referent, i.e., as expressing information which is relevant to and which increases the addressee’s knowledge of this referent (Lambrecht 1994: 131).
A constituent is a topic expression if the proposition expressed by the clause with which it is associated is pragmatically construed as being about the referent of this constituent (Lambrecht 1994: 131).
The comment is what is predicated about the topic. The comment is new in relation to the topic in the sense that it is new information that is asserted, questioned, etc., about the topic (adapted from Gundel and Fretheim 2004).
Clauses were therefore classified as categorical, i.e., as composed of an overt topic constituent and comment, if the clause predicates something about an entity or event that is referred to by means of an overt linguistic constituent in the clause. If, on the other hand, the clause is to be seen as an unanalyzed block that can only be considered to predicate something about circumstances found outside the clausal domain, it is classified as thetic or as all-comment (cf. Section 2.1). Tokens of the eight constructions under investigation were therefore analyzed for the – intuitively identifiable – absence or presence of an aboutness relationship within the confines of the clausal domain. Following Lambrecht (1994), we also allowed in our classification for sentences without any aboutness relationship at all, viz., sentences that are neither categorical nor thetic because they do not have an overt topic constituent and do not predicate something about circumstances found outside the clausal domain. This is the case for so-called ‘identificational sentences’, which serve to identify a referent as a missing argument in an open proposition. Although the referent adds relationally new information to the open proposition and although the relationship in identificational sentences thus resembles to some degree the relationship found in overt topic-constituent-comment sentences, this relationship is still different from a full-fledged overt topic constituent-comment relationship, as the open proposition is referentially incomplete and syntactically not a constituent (cf. Lambrecht 1994: 122 for a discussion).
Based on the distinction between focused constituents and backgrounded constituents, the eight structures under investigation were classified according to Lambrecht’s (1987, 1994, 2000, 2001 typology of the three focus types, viz., argument-focus (with only an argument constituent in focus), predicate-focus (when the subject is backgrounded, while [some part of] the predicate is focused), and sentence-focus (when both the subject and the predicate constituent are focused, with other elements that are being focused or backgrounded not having a direct impact on the qualification). To assess the focus-background articulation of utterances, one needs an explicit definition of the key notion of focus. Given that it is Lambrecht’s notion of sentence-focus articulation that this article aims to investigate, we adopted his notion of focus, which is defined as follows:
The semantic component of a pragmatically structured proposition whereby the assertion differs from the presupposition. (Lambrecht 1994: 213)
For Lambrecht, the key concepts of assertion and presupposition need to be understood by means of reference to the common ground between interlocutors (Stalnaker 1999) rather than by reference to a necessary logical precondition of existence as in the classical logical approach to assertion and presupposition (cf. Strawson 1950). It is important to note that Lambrecht’s specific definition of focus is not necessarily incompatible with more formalized definitions, most notably the “indicator of alternatives” favored in formal semantics theories such as structured meanings (Krifka 2008), although the definitions are not identical (cf. Matić 2015 for a discussion).
4 The multifunctionality of the Italian thetic and sentence-focus constructions
4.1 Syntactic Inversion Construction
In addition to its thetic all-comment and sentence-focus uses (Section 2.2.2), there are several uses of the SIC that are incompatible with the hypothesis of a grammatically encoded thetic and sentence-focus semantics. In Italian, the SIC can correspond to radically different interpretations, as shown by the fact that the same SIC token can be a felicitous answer to different questions, as in, e.g., (61) and (62). The question in (61) elicits a thetic and sentence-focus reading of the sentence; in contrast, the question in (62) elicits a reply with argument-focus construal on the morphosyntactic subject Carla and therefore a relationship of identification between an open proposition (‘X arrived’) and the variable filling in this open proposition. Therefore, it is easily possible to find instances of the SIC that are characterized by argument-focus construal, whereby only the morphosyntactic subject is in focus and the predicate is backgrounded (Belletti 2004, 2018; Bernini 1995; Berretta 1995; Cardinaletti 2018; Meulleman 2012; Sasse 1995, 2006; Sornicola 1995; Wandruszka 1982).
In Examples (62), (63) and (64), it is only the morphosyntactic subject at the end of the sentence, i.e., Carla, tua madre, and Gianni, that is focused, whereas the other constituents of the clauses, in particular the VP, i.e., è arrivata, ha telefonato, and ha parlato, are presupposed in the context. This can be demonstrated by the fact that they can, in principle, be felicitously left out in the reply. On the level of topic-comment structure, none of the three aforementioned examples can be said to constitute a single predication about something outside the clausal domain, so these sentences cannot be analyzed as thetic or all-comment. At the same time, it has to be noted that the relation that holds between the three open propositions and the three filled-in variables, i.e., the three subjects in (62), (63) and (64), is not exactly a relationship of aboutness but rather a relationship of identification. This entails that in these sentences, which are neither thetic nor categorical, no topic is present and that they can be properly understood only in terms of their focus-background articulation rather than in terms of their topic-comment structure (cf. Lambrecht 1994 for a discussion).
A potential issue for our analysis is that it has been argued that the SIC with argument-focus and the SIC with sentence-focus might correspond to two, formally slightly different, constructions in their own right. In fact, Belletti (2004, 2005, 2018) has argued that, following the theoretical assumptions of syntactic Cartography, syntactic differences need to be involved, as in her analysis of the SIC with argument-focus, only the postverbal subject occupies the low FocP, whereas in the SIC with sentence-focus, it is the VP that occupies the low FocP. Although Belletti proposes that argument-focus SIC and sentence-focus SIC reflect two different structures, the specific formal properties actually distinguishing these two types of SIC have yet to be pointed out. In addition, it must be noted that even with the Cartographic framework, Belletti’s hypothesis is the object of considerable controversy. For example, Cardinaletti (2018) has recently argued that the subject position in argument-focus SIC and sentence-focus SIC is essentially the same and that in these cases, there is no “one to one correlation between interpretation and syntactic structure” (Cardinaletti 2018: 80). Furthermore, both Belletti (2018: 52, fn.5) and Cardinaletti (2018) observe that the prosodic difference between sentence-focus SIC and argument-focus SIC is not directly evident (pace Wehr 2012). This observation is in line with the well-known prosodic ambiguity between broad focus and sentence-final narrow focus in statements (Avesani and Vayra 2003; Reinhart 2006; Selkirk 2002), indicating that a prosodic difference between the two SICs cannot be taken for granted (Bernini 1995). Therefore, the prosodic and morphosyntactic differences between SIC with argument-focus and SIC with sentence-focus, if any, seem at the very least still subject to major controversy and in need of further investigation. We maintain that in the absence of any evidence to the contrary, sentence-focus SIC and argument-focus SIC should be considered two information structural uses of one and the same construction.
In contrast to the situation with argument-focus SIC and sentence-focus SIC, there are cases of linear verb-subject order that should be considered formally different constructions from the SIC. This is the case with verb-subject order with topical subjects that are dislocated to the right (Belletti 2018; Bernini 1995; Bonvino 2005; Fiorentino 2005; Wandruszka 1982; Wehr 2012), as in Examples (65), (66) and (67). These constructions convey a topic constituent-comment structure where the predicate, including the verb and, if present, the direct object, needs to be analyzed as the comment and the right dislocated subjects as the overt topic constituent. This can be demonstrated by constructing the appropriate questions that would trigger these sentences as answers. In all three cases, the question would amount to a “What about the subject NP?” question. The right dislocated subject constituent is furthermore presupposed and can in principle be left out without changing the interpretation of the sentence, while the VP is focused, resulting in cases of Lambrechtian predicate-focus construal.
It has been convincingly shown by many authors, including scholars working both inside (Belletti 2004, 2018; Cardinaletti 2002) and outside the framework of Generative Grammar (Bernini 1995; Berruto 1986b; Bonvino 2005; Wehr 2012), that these structures with a postverbal dislocated topical subject are constructions in their own right that should not be subsumed under the SIC. In these cases, the subject is in a syntactic position outside the nuclear clause and is prosodically marked as dislocated by a melodic compression, i.e., a significant reduction in pitch range, occurring after the prosodic focus (Bonvino 2005; Crocco 2013; Crocco and Badan 2020; Wehr 2012).
The SIC remains a very broad and heavily underspecified constructional template that has different characteristics depending on which verbal construction it is combined with and in particular on whether a transitive, unergative or unaccusative verb is involved. Although the instances of the SIC that are specified for the kind of verb differ considerably with regard to their morphosyntax, it is still possible to postulate a more schematic, underspecified constructional template at the basis of all these less schematic, more filled-in constructions. The SIC as such merely stipulates that the verb should precede the subject within the same nuclear clause, leaving room for the insertion of an optional additional object constituent and leaving room for the different syntactic status of the subjects of unergative and unaccusative verbs (cf. Burzio 1986). The more specified and more filled-in constructions with specific verb types are furthermore unlikely to correspond in a one-to-one fashion to a specific information structural function, since all three major formal subtypes of SIC, i.e., transitive SIC, unergative SIC and unaccusative SIC, can be found with predicate-focus, sentence-focus and argument-focus construal (Belletti 2004, 2005, 2018; Berretta 1995; Fiorentino 2005; Meulleman 2012; Venier 2002).
4.2 Existential Construction
Following Cruschina (2012, 2015, 2018), we distinguish the Italian EC proper from the Italian locative construction and its two subtypes (Section 2.2.2). Italian locative constructions are characterized by an identificational relationship and argument-focus construal (Cruschina 2012, 2015, 2018), rather than by the all-comment structure and sentence-focus construal typical of the Italian EC. However, even the Italian EC proper, distinguished from the two Italian locative constructions, allows for a topic constituent-comment structure and predicate-focus construal in addition to its more predominant uses with all-comment structure and sentence-focus construal (Cruschina 2012, 2015, 2018). In fact, the Italian EC can perfectly be used to express a topic constituent-comment structure and predicate-focus construal, as in Examples (68), (69) and (70). In all three examples, the initial locative adverbials nel sistema solare, nell’entrata, and su quest’albero and their coreferential expletive proforms ci have the status of topic constituents, whereas the VP and the pivot NP that follow ci constitute the comment that provides information about the topic constituent. The topical status of the locative adverbials and the coreferential ci becomes clear by the fact that all three examples can figure as the answer to a “What about X?” question that informs about the solar system, the hallway and a particular tree. All three examples can be said to have either a sentence-focus construal or a predicate-focus construal (Cruschina 2012, 2015). In case the locative adverbials are backgrounded, they can be felicitously left out from the examples. In those cases, the expletive proform ci, which functions as a pro-argument in the EC (Burzio 1986; Cruschina 2012, 2015) and which is coreferential with the locative adverbial, needs to be considered as backgrounded as well. The verb and the postverbal NP, on the other hand, are focused, which results in cases of Lambrechtian predicate-focus construal (Cruschina 2012, 2015).
The multifunctionality of the Italian EC is enabled by the fact that the ci in the EC can be used with a host of pragmatic nuances (Cruschina 2018). The various types of ci identified by Berruto (1986a) and De Cesare (2007) can be considered pragmatic uses of ci whose instantiations depend on the specific context involved (Cruschina 2018). In particular, the various uses of ci range from indicating a very abstract location to indicating more specific locations. In the latter case, one can find uses of the EC characterized by a topic constituent-comment structure and predicate-focus construal (Cruschina 2012, 2015, 2018). The information structural multifunctionality of the Italian EC and the strong link between its information structural articulation and the pragmatic interpretation of the ci introducing the EC is strikingly similar to the situation in Dutch, where a similar link exists between the predicate-focus and sentence-focus articulation of the Dutch EC and the pragmatic interpretation of the proform er introducing these clausal constructions (cf. Belligh 2020a, 2020b).
4.3 Presentational Cleft
The PC has been considered a very prototypical thetic and sentence-focus construction, even to the extent that semantic traits associated with theticity and sentence-focus are sometimes included among the defining traits of the PC. In particular, the requirement of a “semantically empty” ci has been proposed as part of the definition of the PC (Section 2.2.3). However, recent corpus research that does not a priori exclude certain PCs on semantic and pragmatic grounds has revealed uses of this construction that diverge strongly from theticity or sentence-focus (Karssenberg et al. 2017). In particular, it is possible to find uses of the PC with argument-focus construal and an identificational relation in the sense that the construction conveys an open proposition that is filled by a missing variable, e.g., Examples (71), (72), and (73). Only the NP that follows the inflected instances of the copula essere is focused, i.e., la giunta, il PM, and Maria, whereas the preceding c’è and the ensuing subordinate clause are presupposed and backgrounded.
Furthermore, it is possible to find uses of the PC with a topic constituent-comment structure and predicate-focus construal in which the sentence initial ci refers to a situationally, textually or inferentially accessible location and functions as the topic constituent of the clause. This is very similar to the topic constituent-comment and predicate-focus uses of the EC, as in, e.g., Examples (74) and (75). In these two cases, the clause-initial ci is coreferential with ‘the table’ and the ‘east of Sicily’, respectively, which can both be interpreted as the entities that the clauses are providing information about. This can be demonstrated by the fact that both examples can be used as answers to “What about X?” questions.
A possible issue for our analysis is that it has been argued that the various information structural uses identified above betoken formally different constructions. In particular, it has been argued that the argument-focus and predicate-focus construals of the PC are actually instances of a locative construction that is accompanied by an additional relative clause, either restrictive or appositive, and an EC accompanied by an additional relative clause, either restrictive or appositive, respectively (Cruschina 2012, 2015, 2018). The formal differences between these structures that are at first glance identical would lie in the fact that the PC proper involves a pseudorelative clause, whereas both the EC and the locative constructions are always used with proper relative clauses. By adopting a very narrow definition of the PC with the necessary presence of a pseudorelative rather than a proper relative, PCs could be said to be dedicated to the expression of theticity and sentence-focus construal.
However, to be relevant for the line of research pursued here, there needs to be a one-to-one relationship between form and function. The issue therefore hinges on the fact that there is truly a one-to-one correlation between the kind of relative clause and the information structural articulation of biclausal ci sentences. On the basis of the currently available evidence, this seems to be an unwarranted claim. First, it must be noted that Cruschina himself (2012, 2015) is very cautious in his formulation, stressing at various points that the PC “typically” occurs with a pseudorelative. Again, for the line of investigation pursued here, there must be a one-to-one correlation. Second, the ci sentences cited above clearly have argument-focus and predicate-focus construal, while their relative clauses have various characteristics that are typically ascribed to the pseudorelative clause (cf. Section 2.2.4). In fact, in Examples (71), (72), (73), (74), and (75), the che cannot be replaced by il quale without changing the meaning and nature of the construction involved. Furthermore, in Examples (71), (72), (73), (74), and (75), the complementizer che is the syntactic subject of the subordinate clause. Finally, in Examples (71), (72), (73), and (75), the events described in the main clause and in the relative clause take place simultaneously. Third, it has been pointed out by Marzo and Crocco (2015) that the distinction between a pseudorelative and restrictive relative clause can be highly problematic in cases where the antecedent of the subordinate clause is indefinite. As a result, the kind of neat formal separation between the three types of ci sentences that is necessary to determine a one-to-one relation between form and function is at the very least difficult to establish in a convincing way.
4.4 Avere Cleft
The AC can be used to express several functions that are established negatively in relation to the categories theticity and sentence-focus. In particular, the AC can be used to express argument-focus, with a (constrastive) focus reading on the morphosyntactic direct object, as in, e.g., (76), or with a (constrastive) predicate-focus on the predicate of the relative clause, as in, e.g., (77). For example, in (76), only the direct object NP la testa is focused, whereas the fact that something hurts is presupposed as is the fact that someone assumed that it was the speaker’s back that was aching. In (77), by contrast, the predicate of the relative clause is (contrastively) focused, whereas the fact that something is going on with the speaker’s head is already presupposed.
Additionally, the same syntactic configuration that formally defines the AC can in principle be used to convey a topic constituent-comment structure with predicate-focus construal, as in, e.g., (78). The fact that the speaker has a headache is constructed in this example as a piece of relevant new information about the speaker, who is grammatically evoked only by the verb that is inflected for the first-person singular. The subject of the predication is also clearly presupposed, which is demonstrated by the fact that it is not explicitly present in the utterance, except for the grammatical inflection of the main verb, while the verb, the direct object NP and the relative clause are explicitly present and focused.
4.5 Ecco Cleft
The EccoC can be used not only with its thetic and sentence-focus readings but also with readings that are incompatible with the hypothesis of the categories of theticity and sentence-focus as structurally encoded functions. First, the EccoC can be used with focalizers that severely narrow down the focus scope, as in (79). In (79), the object NP la luce is preceded by the focalizing adverb anche ‘too’, scoping on the object NP, thereby inducing a narrow argument-focus reading.
Furthermore, the EccoC also seems to be available for the expression of predicate-focus construal and an overt topic constituent-comment structure. This can be shown by the acceptability of the combination of the EccoC and a construction with the topical right dislocation of the morphosyntactic object, as in, e.g., (80).
4.6 Perception Verb Construction
The PVC allows for various uses that are incompatible with a structurally encoded theticity or sentence-focus semantics. First, the PVC allows for the insertion of adverbs such as proprio ‘really’ and soltanto ‘only’ that force a narrow (contrastive) focus reading of the PVC. This results in sentences with argument-focus construal, as in, e.g., (81) and (82). Both examples are modifications of the original example proposed by Berretta (1995). In (81), proprio ‘really’ or ‘precisely’ takes scope on ‘Maria’, implying that the speaker saw her and no one else, while both the fact that the speaker saw someone and the fact that someone was chatting with friends can be considered to be presupposed. In (82), we can find the same interpretation as in (81), with ‘only’ focusing on Maria, while ‘only’ can also be construed as taking scope on both Maria and the subsequent relative clause. In both cases, the presence of the adverb narrows down the focus scope and hinders a sentence-focus construal.
Second, the PVC can, almost by its very nature, be used for topic constituent-comment and predicate-focus purposes. Since the construction formally consists of a first-person subject, an inflected verb, a NP and then a subordinated clause containing new information, the topic constituent-comment and predicate-focus uses come quite naturally. In fact, the thetic and sentence-focus uses can be read into the construction only if the main verb of the construction is not interpreted with its full lexical meaning but rather in a weak sense as a parenthetical verb (cf. Willems and Blanche-Benveniste 2014). The same verbs that enter in the construction as weak verbs to express theticity and sentence-focus can also be found in the same construction as full lexical verbs that express information about what a certain perceiver is perceiving, as in, e.g., (83). The fact that the speaker sees the soldiers loading the truck is construed here as a piece of new information about the speaker, who is grammatically evoked only by the verb inflected for the first-person singular. The subject of the predication is furthermore presupposed, which is demonstrated by the fact that it is not explicitly present in the utterance, while the verb, the direct object NP and the ensuing relative clause are not presupposed but focused and, hence, cannot be left out.
4.7 Bare NP cleft
The BNPC cannot only be used to convey a thetic and sentence-focus utterance, as in Examples (44) and (45) discussed above, but can also be used for diametrically opposed readings. In fact, the BNPC can be manipulated to express at least a form of argument-focus construal, as in the constructed Example (84). In this case, the specific reading with the focus only on mica is forced by the use of the presuppositional negator mica (lit. ‘crumb’, from Latin mica). The negator takes scope over the rest of the sentence, which is presented as backgrounded and presupposed information, and mica then denies this presupposition as a whole (Cinque 1976). Such a counterexpectational negation is prosodically marked as a (contrastive) focus on mica alone (cf. Magistro et al. forthcoming), which therefore definitely excludes a sentence-focus reading.
Given that the BNPC can be used for readings that are diametrically opposed to theticity and sentence-focus, it cannot be considered a construction that is structurally dedicated to the expression of either of these two functional categories.
4.8 Prototypical Cleft
While prototypical clefts can occasionally be used to express a thetic and sentence-focus reading (Berretta 1995; Roggia 2009), they are predominantly used as a means to express argument-focus construal (Lambrecht 1994, 2001; Roggia 2009), as in (85) and (86). In these cases, it is the clefted NPs Eleonora and questo that are focused, whereas the fact that the speaker can see someone in a company in Milan and the fact that he/she could not find a particular item is presupposed.
These sentences are characterized by argument-focus construal and an identificational relationship between a proposition with an open variable and the information filling the gap in the proposition (cf. Roggia 2009: Ch. 3 for a discussion). Given that the PrC is predominantly used for argument-focus construal and that the sentence-focus and thetic uses of this construction are only marginal uses that, to the best of our knowledge, have hitherto been identified only by Berretta (1995) and Roggia (2009), it should be clear that the PrC is not a construction that can be considered as having theticity or sentence-focus as its structurally encoded semantics.
4.9 Interim conclusion
On the basis of the investigation of the information structural multifunctionality of the eight Italian thetic and sentence-focus constructions that have hitherto been identified in the literature, it turns out that none of these constructions can be considered to grammatically encode theticity or sentence-focus. In fact, all eight constructions can be used to convey uses that are radically opposed to the very notions of theticity and sentence-focus. For reasons specified in the methodology section, it is therefore highly unlikely that these constructions grammatically encode theticity or sentence-focus.
Given that this is true for every Italian construction that has hitherto been identified in the literature as a thetic or sentence-focus construction, one can make the case that Italian does not have theticity and sentence-focus among its grammatically encoded categories. At the same time, it must be noted that theticity and sentence-focus are, at first glance, frequently recurrent uses of several of the eight Italian constructions. Therefore, they are in any case important categories of language use that capture crucial aspects of the pragmatics of several of the investigated Italian constructions. Furthermore, it could in principle turn out to be the case that there are additional, previously unnoted Italian structures that can be used for the expression of theticity and sentence-focus and that might encode theticity or sentence-focus in a grammatical fashion. Before we move on to the overall conclusion of this article, there are three possible objections against the proposed analysis that need to be addressed first.
The first objection pertains to the theoretical assumption that the various uses of the constructions under scrutiny are in fact several uses of one and the same construction. Alternatively, it is possible to maintain that the uses of the various constructions illustrated in the previous sections qualify as the structurally encoded meanings of several homonymous constructions rather than as different pragmatic uses of one and the same construction. In fact, this is the approach adopted by Lambrecht (1987, 1994, 2000, 2001 to account for the various focus articulations that many sentence-focus constructions can exhibit crosslinguistically (cf. Section 2.5). While this alternative strand of analysis is, logically speaking, sound, it has three major disadvantages when compared to the analysis put forward in this article. First, homonymy is, generally speaking, a solution that should be used with caution. If used in an unconstrained fashion, the adoption of homonymous analyses leads to a uncontrolled proliferation of new constructions, viz., a new construction for every single use that can be identified by linguists (Lyons 1977). If one follows Lambrecht’s homonymy solution, the analysis of the grammatical system of Italian would be characterized by rampant homonymy. In addition to the homonymy between sentence-focus and argument-focus constructions acknowledged and discussed by Lambrecht (1987, 1994, 2000, our data indicate that there are also various Italian constructions with both predicate-focus and sentence-focus construal. To account for this multifunctionality in terms of homonymy would entail that one would need to accept the existence of homonymous constructions for each of the three Lambrechtian focus categories, which are negatively established in relation to one another. Second, homonymy between sentence-focus and predicate-focus constructions is a form of homonymy that Lambrecht himself explicitly excludes and emphatically does not want to entertain in his system of focus categories (Section 2.5). It therefore hardly seems a solution to adopt once more a homonymy analysis to account for the fact that a single construction can display both sentence-focus and predicate-focus construal. Third, the very raison d’être of the system of focus categories proposed by Lambrecht is to provide a better alternative for the category of theticity at the level of the grammatically encoded functions of information structure constructions. Precisely because theticity was considered not to be sufficiently reflected in grammatical form, sentence-focus was proposed as an alternative. If we are forced to accept rampant homonymy to make this solution plausible, the sentence-focus approach turns out to be hardly any better than the original thetic approach it seeks to supplant.
The second possible objection relates to the possibility that some of the constructions under scrutiny could be clusters of superficially similar but syntactically quite different constructions, each with its specific interpretative properties (cf. Belletti 2004, 2005, 2018; Cruschina 2012, 2018). In particular, for three of the constructions under scrutiny here, i.e., the SIC, the EC and the PC, it has been argued that they might be better analyzed as several constructions, each with its own specific morphosyntactic make-up and information structural characteristics. In the discussion of the three relevant constructions, we adduced several arguments to support our analysis that we are in fact dealing with one and the same construction in all three cases. The EC did not pose any problems for our proposed analysis, as we focused on the narrowly defined EC that was already distinguished from other types of similar ci sentences on the basis of formal syntax studies (Cruschina 2012, 2018). For the SIC, it became clear that there are currently no convincing arguments to assume that argument-focus SIC and sentence-focus SIC would betoken two formally different constructions in their own right. For the PC, we posited on the basis of several arguments that there is no convincing evidence to accept that the type of subordinate clause involved, i.e., a pseudorelative or a proper relative, would correspond structurally in a one-to-one fashion to a specific information structural articulation.
The third possible objection relates to the fact that in this study we have focused on eight morphosyntactic constructions in Italian largely in isolation from their prosodic characteristics. This leaves out an important part of the picture, as it could turn out to be the case that the various uses of several constructions correspond to distinct forms at the level of prosody. This is not only potentially the case for the three constructions discussed with regard to their possible morphosyntactic differences, but it could in principle be the case for all eight constructions. Given that prosodic characteristics are allowed to qualify as defining traits for constructionhood at the formal level (Goldberg 2006; Hoffman and Trousdale 2013; Lambrecht 1994), we might in principle be dealing with several independent constructions that are identical only with regard to their morphosyntactic properties. At the same time, it has often been noted that even prosodic prominence can be ambiguous with regard to the kind of focus-background articulation it corresponds to. In particular, it has been shown for canonical SVO sentences that if the clause-final constituent is prosodically prominent, the extension of the focus domain can be limited to the final constituent, can apply to the predicate containing it, or can even encompass the whole sentence (Reinhart 2006). Pursuing this important issue fully would require an in-depth investigation of the prosodic characteristics of all eight Italian constructions in the relevant contexts. Within the limits of the present article, this could not be pursued, but we want to highlight the importance of this endeavor for future research.
This article aimed to examine whether theticity and sentence-focus can be considered grammatically encoded categories of Italian. Drawing on Integral Linguistics, we defined grammatically encoded categories as functional categories that have a structurally corresponding counterpart in linguistic form and that are therefore encoded as the indefeasible semantics of at least one construction of a given language. By relying on this conception of grammatically encoded categories, we outlined a methodology for determining whether a language has a certain functional category as a grammatically encoded category. In particular, the method consists of first identifying the relevant constructions that can be used to convey a certain functional category (the onomasiological perspective). Subsequently, each of the identified constructions needs to be studied in its own right (the semasiological perspective) to investigate whether its various uses can be aligned with the hypothesis of the functional category under investigation as the grammatically encoded meaning or function of the construction. The article then focused on the eight Italian thetic and sentence-focus constructions that have hitherto been identified in the literature. We examined whether these constructions displayed information structural uses that could falsify the hypothesis of theticity or sentence-focus being the structurally encoded nondenotational semantics of these constructions. In particular, we investigated whether the eight Italian constructions under scrutiny could also be used to convey diametrically opposed information structural uses, such as argument-focus and predicate-focus construal and an overt topic constituent-comment structure.
Our analysis supports the view that theticity and sentence-focus are not structurally encoded in any of the Italian constructions, as the eight examined structures can also be used to convey topic constituent-comment structures, argument-focus construal and predicate-focus construal. At the same time, we acknowledged a number of possible objections against the proposed analysis. In particular, we discussed the homonymy alternative proposed by Lambrecht (1987, 1994, 2000, 2001 as being logically sound but connected with several major disadvantages at the theoretical level. In addition, we entertained the possibility that we might be dealing with several constructions with their own formal and information structural traits that are only superficially similar but adduced various arguments against this way of analyzing the structures involved. Finally, we acknowledged not properly taking into account the link between the prosodic features of the constructions and their information structural uses as an important limitation of this study while highlighting the importance of pursuing this line of research for future work.
Overall, our results indicate that theticity and sentence-focus are not grammatically encoded categories of Italian but rather categories of language use. However, to substantiate this claim even more fully, it is necessary to investigate every possible Italian construction that can be used to convey theticity and sentence-focus. While to our knowledge, we examined all Italian syntactic structures that have been identified in the literature as thetic and sentence-focus constructions, this does not guarantee that we in fact demonstrated the multifunctionality of every single thetic and sentence-focus construction in Italian. If future research identifies additional Italian thetic and sentence-focus constructions, these constructions need to be investigated as well with regard to their multifunctionality to corroborate or falsify our proposed conclusion.
We surmise that the fact that none of the eight investigated Italian thetic and sentence-focus constructions is structurally dedicated to the expression of theticity and sentence-focus might provide some answers as to why there are many thetic and sentence-focus constructions in Italian. The literature on theticity and sentence-focus has been dominated by the onomasiological point of view and has therefore quite liberally identified all structures that can potentially be used for the expression of theticity and sentence-focus as thetic and sentence-focus constructions. As noted by Lambrecht (1994, 2001), language users are usually more than happy to use the available constructions of a language for all kinds of purposes. Given that there are only a limited number of relevant major categories at the level of relational givenness, viz., all-comment, topic constituent-comment, predicate-focus, argument-focus and sentence-focus (Lambrecht 1994), and given that the number of possible sentential structures in Italian is considerably higher, it is understandable that for the expression of one specific information structural function, there are several linguistic structures available.
In addition to being relevant for furthering our understanding of the various Italian thetic and sentence-focus constructions, our findings also contribute to the general discussion of the grammatical or general conceptual status of the categories of theticity and sentence-focus. The present article can only make the case that the two categories are not grammatically encoded in Italian. In fact, it is perfectly possible that some natural language grammatically encodes one or both of these two functional categories. It is therefore highly relevant to study thetic and sentence-focus constructions in as many languages as possible to cast further light on this issue. However, if we combine the insights offered in this article with the existing insights found in the literature, an interesting pattern emerges. Theticity has been conceptualized from its origins (Marty 1918) as a logical category that is not necessarily reflected in linguistic form (cf. Belligh 2020b; Sornicola 1995; Ulrich 1985; Venier 2002). Over the past 25 years, several thetic and sentence-focus constructions in various languages have been shown to display broad multifunctionality that is incompatible with the hypothesis of a structurally encoded semantics of theticity or sentence-focus (Belligh 2020a; Karssenberg 2016; Karssenberg and Lahousse 2018; Karssenberg et al. 2017; Matić 2003; Sasse 1995, 2006). To our knowledge, Italian (the language investigated in this article) and Dutch (Belligh 2020b) are the only two languages that have hitherto been examined with regard to all their previously identified thetic and sentence-focus constructions, and in both languages theticity and sentence-focus turn out to be not grammatically encoded by the constructions under scrutiny. Judging on the basis of the currently available evidence, it could be that theticity and sentence-focus may not be grammatically encoded in general but rather should be considered categories of language use across the board. However, much more research on a wide array of typologically different languages is needed to make any substantial claims about the overall nature of the categories of theticity and sentence-focus.
In regard to the theoretical question pertaining to the position of information structure phenomena on the grammar-semantics-pragmatics interface, our results are in line with the position adopted by Féry (2008) that information structural categories are not structurally encoded on a one-to-one basis by specific linguistic structures. Our analysis is furthermore to a large degree compatible with the analysis of Matić and Wedgwood (2013), who show that crosslinguistically, many focus structures are not systematically dedicated to the expression of a universal category of focus. At the same time, it has to be noted from the perspective of Integral Linguistics that, even if theticity and sentence-focus turn out not to be grammatically encoded categories for an ever-increasing number of languages, this does not entail that they are no longer meaningful functional categories of language. Theticity, as has been argued from its inception, is a relevant property of propositions that can be conveyed without necessarily relying on structurally dedicated form-meaning pairings in any language system. The focus categories of Lambrecht, which are defined on the basis of a Stalnakerian account of discourse, remain in any case highly relevant to understanding the nature of discourse. These categories can interact with linguistic structure without necessarily having to be structurally encoded in the grammars of specific languages, although they were initially devised by Lambrecht to fulfil that function. Functional categories that are not grammatically encoded in a linguistic system are not merely “comparative concepts” (Haspelmath 2010) or “heuristic tools” (Matić and Wedgwood 2013) but can be conveyed and conceptualized by language users, who always go beyond the limited list of categories that are grammatically encoded in the systems of the languages that they speak (Coseriu 1975 , 1989, 1992 , 2007 ; Willems 1997, 2011, 2016; Willems and Munteanu 2021).
Abraham, Werner, Elisabeth Leiss & Yasuhiro Fujinawa (eds.). 2020. Thetics and categoricals (Linguistik Aktuell/Linguistics Today 262). Amsterdam & Philadelphia: John Benjamins.10.1075/la.262Search in Google Scholar
Avesani, Cinzia & Mario Vayra. 2003. Broad, narrow and contrastive focus in Florentine Italian. In Maria-Josep Sole, Daniel Recasens & Joaquin Romero (eds.), Proceedings of the 15th international congress of phonetic sciences, 1803–1806. Barcelona: Causal Productions.Search in Google Scholar
Belletti, Adriana. 2004. Aspects of the low IP area. In Luigi Rizzi (ed.), The structure of CP and IP: The cartography of syntactic structures, vol. 2, 16–51. Oxford: Oxford University Press.Search in Google Scholar
Belletti, Adriana. 2018. Revisiting the cartography of (Italian) postverbal subjects from different angles with reference to canonicality considerations. Italian Journal of Linguistics 30. 37–58.Search in Google Scholar
Belligh, Thomas. 2018. The role of referential givenness in Dutch alternating presentational constructions. Belgian Journal of Linguistics 32. 21–52. https://doi.org/10.1075/bjl.00015.bel.Search in Google Scholar
Belligh, Thomas. 2020a. Dutch thetic and sentence-focus constructions on the semantics-pragmatics interface: A case study. Studies in Language 44(4). 831–878. https://doi.org/10.1075/sl.19021.bel.Search in Google Scholar
Belligh, Thomas. 2020b. Are theticity and sentence-focus encoded grammatical categories of Dutch? In Werner Abraham, Elisabeth Leiss & Yasuhiro Fujinawa (eds.), Thetics and categoricals (Linguistik Aktuell/Linguistics Today 262), 33–68. Amsterdam & Philadelphia: John Benjamins.10.1075/la.262.02belSearch in Google Scholar
Belligh, Thomas & Klaas Willems. 2021. What’s in a code? The code-inference distinction in Neo-Gricean Pragmatics, Relevance Theory, and Integral Linguistics. Language Sciences 83. https://doi.org/10.1016/j.langsci.2020.101310.Search in Google Scholar
Bellini, Daniele & Stefan Schneider. 2003–2019. Banca dati dell’italiano parlato (BADIP). Graz: Karl-Franzens-Universität Graz.Search in Google Scholar
Bentley, Delia, Francesco Maria Ciconte & Silvio Cruschina. 2015. Existentials and locatives in Romance dialects of Italy. Oxford: Oxford University Press.10.1093/acprof:oso/9780198745266.001.0001Search in Google Scholar
Bernini, Giuliano. 1995. Verb-subject order in Italian: An investigation of short announcements and telecast news. Sprachtypologie und Universalienforschung 48. 44–71. https://doi.org/10.1524/stuf.1918.104.22.168.Search in Google Scholar
Berretta, Monica. 1995. Come inseriamo elementi nuovi nel discorso/1: ‘C’è il gatto che ha fame’. Italiano e oltre 53. 79–105.Search in Google Scholar
Berruto, Gaetano. 1986a. Un tratto sintattico dell’italiano parlato: Il c’è presentativo. In Klaus Lichem, Edith Mara & Susanne Knaller (eds.), Parallela 2: Aspetti della sintassi dell’italiano contemporaneo, 61–73. Tübingen: Gunter Narr.Search in Google Scholar
Berruto, Gaetano. 1986b. Le dislocazioni a destra in italiano. In Harro Stammerjohann (ed.), Tema-rema in italiano/Theme-rheme in Italian/Thema-rhema im Italienischen, 55–69. Tübingen: Gunter Narr.Search in Google Scholar
Bonvino, Elisabetta. 2005. Le sujet postverbal en italien parlé: Syntaxe, zones et intonation. Paris: Ophrys.Search in Google Scholar
Cardinaletti, Anna. 2018. On different types of postverbal subjects in Italian. Italian Journal of Linguistics 30(2). 79–106.Search in Google Scholar
Casalicchio, Jan. 2013. Pseudorelative, gerundi e infiniti nelle varietà romanze: Somiglianze (solo) superficiali e corrispondenze strutturali. Munich: LINCOM.Search in Google Scholar
Cinque, Guglielmo. 1976. Mica. Annali della facoltà di lettere e filosofia, università di Padova I, 101–112. Padova: University of Padova Press.Search in Google Scholar
Cinque, Guglielmo. 1988. La frase relativa. In Lorenzo Renzi (ed.), Grande grammatica italiana di consultazione, vol. I, 443–503. Bologna: il Mulino.Search in Google Scholar
Coene, Ann. 2006. Lexikalische Bedeutung, Valenz und Koerzion. Hildesheim: Olms.Search in Google Scholar
Coene, Ann & Klaas Willems. 2006. Konstruktionelle Bedeutungen: Kritische Anmerkungen zu Adele Goldbergs Konstruktionsgrammatischer Bedeutungstheorie. Sprachtheorie und Germanistische Linguistik 16. 1–35.Search in Google Scholar
Coseriu, Eugenio. 1975 . Sprachtheorie und allgemeine Sprachwissenschaft. 5 Studien. Munich: Wilhelm Fink Verlag.Search in Google Scholar
Coseriu, Eugenio. 1987. Formen und Funktionen. Studien zur Grammatik. Tübingen: Max Niemeyer.Search in Google Scholar
Coseriu, Eugenio. 1992 . Einführung in die Allgemeine Sprachwissenschaft. Tübingen: Francke.Search in Google Scholar
Coseriu, Eugenio. 2007 . Sprachkompetenz. Grundzüge der Theorie des Sprechens. Tübingen: Gunter Narr Verlag.Search in Google Scholar
Crocco, Claudia. 2013. Is Italian clitic right dislocation grammaticalised? A prosodic analysis of yes/no questions and statements. Lingua 133. 30–52. https://doi.org/10.1016/j.lingua.2013.03.003.Search in Google Scholar
Cruschina, Silvio. 2012. Focus in existential sentences. In Valentina Bianchi & Cristiano Chesi (eds.), Enjoy linguistics! Papers offered to Luigi Rizzi on the occasion of his 60th birthday, 77–107. Siena: CISCL Press.Search in Google Scholar
Cruschina, Silvio. 2015. Focus structure. In Delia Bentley, Francesco Ciconte & Silvio Cruschina (eds.), Existentials and locatives in romance dialects of Italy, 43–98. Oxford: Oxford University Press.10.1093/acprof:oso/9780198745266.003.0002Search in Google Scholar
De Cesare, Anna-Maria. 2007. Sul cosidetto ‘c’è presentativo’. Forme e funzioni. In Anna-Maria De Cesare & Angela Ferrari (eds.), Lessico, grammatica e testualità, tra italiano scritto e parlato, 127–153. Basel: University of Basel Press.Search in Google Scholar
Deguchi, Masanori. 2012. Revisiting the thetic/categorical distinction in Japanese. Poznań Studies in Contemporary Linguistics 48(2). 223–237. https://doi.org/10.1515/psicl-2012-0011.Search in Google Scholar
De Mauro, Tullio, Federico Mancini, Massimo Vedovelli & Miriam Voghera. 1993. Lessico di frequenza dell’italiano parlato. Milan: Etaslibri.Search in Google Scholar
Dufter, Andreas & Christoph Gabriel. 2016. Information structure, prosody, and word order. In Susann Fischer & Christoph Gabriel (eds.), Manual of grammatical interfaces in Romance, 419–455. Berlin & Boston: De Gruyter Mouton.10.1515/9783110311860-017Search in Google Scholar
Fiorentino, Giuliana. 1999. Relativa debole: Sintassi, uso, storia in italiano. Milan: Franco Angeli.Search in Google Scholar
Fillmore, Charles J., Kay Paul & O’Connor Mary Catherine. 1988. Regularity and idiomaticity in grammatical constructions: The case of let alone. Language. 501–538. https://doi.org/10.2307/414531.Search in Google Scholar
Fillmore, Charles J. & Paul Kay. 1993. Construction Grammar coursebook. Berkeley, CA: University of California, Berkeley Press.Search in Google Scholar
Goldberg, Adele. 1995. Constructions: A Construction Grammar approach to argument structure. Chicago: University of Chicago Press.Search in Google Scholar
Goldberg, Adele. 2006. Constructions at work: The nature of generalization in language. Oxford: Oxford University Press.Search in Google Scholar
Gundel, Jeanette K. 1988 . The role of topic and comment in linguistic theory. New York: Garland.Search in Google Scholar
Gundel, Jeanette K. & Thorstein Fretheim. 2004. Topic and focus. In Lawrence Horn & Gregory Ward (eds.), The Handbook of Pragmatics, 175–196. Malden, MA: Blackwell.10.1002/9780470756959.ch8Search in Google Scholar
Halliday, Michael Alexander Kirkwood & Christian Matthiessen. 2004. An introduction to functional grammar. Oxford: Oxford University Press.Search in Google Scholar
Jackendoff, Ray. 2010. Meaning and the lexicon: The parallel architecture. Oxford: Oxford University Press.Search in Google Scholar
Karssenberg, Lena. 2016. French il y a clefts, existential sentences and the focus-marking hypothesis. Journal of French Language Studies 27. 405–430. https://doi.org/10.1017/s0959269516000296.Search in Google Scholar
Karssenberg, Lena & Karen Lahousse. 2018. The information structure of French il y a clefts and c’est clefts: A corpus-based analysis. Linguistics 56(3). 513–548. https://doi.org/10.1515/ling-2018-0004.Search in Google Scholar
Karssenberg, Lena, Marzo Stefania, Karen Lahousse & Daniela Gugliemo. 2017. There’s more to Italian c’è clefts than expressing all-focus. Italian Journal of Linguistics 29(2). 57–86.Search in Google Scholar
Kuroda, Sige-Yuki. 1972. The categorical and the thetic judgment: Evidence from Japanese syntax. Foundations of Language 9. 153–185.Search in Google Scholar
Lambrecht, Knud. 1987. Sentence-focus, information structure, and the thetic-categorical distinction. Berkeley Linguistics Society 13. 366–382. https://doi.org/10.3765/bls.v13i0.1800.Search in Google Scholar
Leino, Jaakko. 2013. Information structure. In Thomas Hoffmann & Graeme Trousdale (eds.), The Oxford handbook of Construction Grammar, 329–345. Oxford: Oxford University Press.10.1093/oxfordhb/9780195396683.013.0018Search in Google Scholar
Lyons, John. 1977. Semantics. Cambridge: Cambridge University Press.Search in Google Scholar
Magistro, Giuseppe, Claudia Crocco & Anne Breitbarth. forthcoming. Information structure and Jespersen’s cycle: The dialects of Veneto as a window on processes of language change. In Chiara De Bastiani, Marco Coniglio & Nicholas Catasso (eds.), Language change at the interfaces: Intersentential and intrasentential phenomena. Amsterdam & Philadelphia: John Benjamins.10.1075/la.275.01magSearch in Google Scholar
Marty, Anton. 1918. Gesammelte Schriften. Halle: Max Niemeyer.Search in Google Scholar
Matić, Dejan. 2003. Topics, presuppositions, and theticity: An empirical study of verb-subject clauses. Cologne: Universität Köln doctoral dissertation.Search in Google Scholar
Matić, Dejan. 2015. Information structure in linguistics. In James D. Wright (ed.), The international encyclopedia of social and behavioral sciences, 2nd edn., vol. 12, 95–99. Amsterdam: Elsevier.10.1016/B978-0-08-097086-8.53013-XSearch in Google Scholar
Matić, Dejan & Daniel Wedgwood. 2013. The meanings of focus: The significance of an interpretation-based category in cross-linguistic analysis. Journal of Linguistics 49. 127–163.10.1017/S0022226712000345Search in Google Scholar
Meulleman, Machteld. 2012. Les localisateurs dans les constructions existentielles: Approche comparée en espagnol, en français et en italien. Berlin & Boston: De Gruyter Mouton.10.1515/9783110263473Search in Google Scholar
Pinelli, Maria Cristina, Cinzia Avesani & Cecilia Poletto. 2018. Is it prosody that settles the syntactic issue? An analysis of Italian cleft sentences. In Alessandro Vietti, Lorenzo Spreafico, Daniela Mereu & Vincenzo Galatà (eds.), Speech in the natural context, 157–177. Milan: Officina Ventuno.Search in Google Scholar
Radford, Andrew. 1975. Pseudo-relatives and the unity of subject raising. Archivum Linguisticum 6. 32–64.Search in Google Scholar
Roggia, Carlo Enrico. 2009. Le frasi scisse in italiano: Struttura informativa e funzioni discorsive. Geneva: Slatkine.Search in Google Scholar
Sasse, Hans-Jürgen. 1995. ‘Theticity’ and VS order: A case study. In Yaron Matras & Hans-Jürgen Sasse (eds.), Verb-subject order and theticity in European languages, 3–31. Berlin: Akademie-Verlag.10.1524/stuf.1922.214.171.124Search in Google Scholar
Sasse, Hans-Jürgen. 2006. Theticity. In Giuliano Bernini & Marcia Schwartz (eds.), Pragmatic organization of discourse in the languages of Europe, 255–308. Berlin & New York: Mouton de Gruyter.10.1515/9783110892222.255Search in Google Scholar
Selkirk, Elisabeth. 2002. Contrastive FOCUS vs. presentational focus: Prosodic evidence from right node raising in English. In Bernard Bel & Isabel Marlin (eds.), Speech prosody 2002: Proceedings of the first international conference on speech prosody, 643–646. Aix-en-Provence: Laboratoire Parole et Langage.Search in Google Scholar
Selkirk, Elisabeth. 2008. Contrastive focus, givenness and the unmarked status of “discourse-new”. Acta Linguistica Hungarica 55. 331–346. https://doi.org/10.1556/aling.55.2008.3-4.8.Search in Google Scholar
Sornicola, Rossanna. 1995. Theticity, VS order and the interplay of syntax, semantics and pragmatics. Sprachtypologie und Universalienforschung 48. 72–83. https://doi.org/10.1524/stuf.19126.96.36.199.Search in Google Scholar
Ulrich, Miorita. 1985. Thetisch und kategorisch: Funktionen der Anordnung von Satzkonstituenten am Beispiel des Rumänischen und anderer Sprachen. Tübingen: Gunter Narr.Search in Google Scholar
Van der Gucht, Fieke, Klaas Willems & Ludovic De Cuypere. 2007. The iconicity of embodied meaning. Polysemy of spatial prepositions. Language Sciences 29(6). 733–754. https://doi.org/10.1016/j.langsci.2006.12.027.Search in Google Scholar
Venier, Federica. 2002. La presentatività. Sulle tracce di una nozione. Alessandria: Edizioni dell’Orso.Search in Google Scholar
Wandruszka, Ulrich. 1982. Studien zur italienischen Wortstellung. Tübingen: Gunter Narr.Search in Google Scholar
Wehr, Barbara. 2012. Syntax und Prosodie: V–S im Neuitalienischen. In Elmar Schafroth & Maria Selig (eds.), Testo e ritmi. Zum Rhythmus in der italienischen Sprache, 205–222. Frankfurt: Peter Lang.Search in Google Scholar
Willems, Dominique & Claire Blanche-Benveniste. 2014. A constructional corpus-based approach of ‘weak’ verbs in French. In Hans Boas & Francisco Gonzálvez-García (eds.), Romance perspectives on Construction Grammar, 113–138. Amsterdam & Philadelphia: John Benjamins.10.1075/cal.15.04wilSearch in Google Scholar
Willems, Klaas. 1997. Kasus, grammatische Bedeutung und kognitive Linguistik: Ein Beitrag zur allgemeinen Sprachwissenschaft. Tübingen: Gunter Narr.Search in Google Scholar
Willems, Klaas. 2011. Meaning and interpretation: The semiotic similarities and differences between cognitive grammar and European structural linguistics. Semiotica 2011(185). 1–50. https://doi.org/10.1515/semi.2011.032.Search in Google Scholar
Willems, Klaas. 2016. The universality of categories and meaning: A Coserian perspective. Acta Linguistica Hafniensa 48(1). 110–133. https://doi.org/10.1080/03740463.2016.1141565.Search in Google Scholar
© 2022 Thomas Belligh and Claudia Crocco, published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.